Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    file format software language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ሽፋን
    0.70
     بیاکتنې
    0.68
     ሂደት
    0.67
     የተለያዩ
    0.66
     ప్రభావ
    0.65
     என்பதை
    0.64
     onslaught
    0.64
     المثلثيه
    0.64
    烈的
    0.63
     വ്യാപ
    0.63
    POSITIVE LOGITS
    o
    0.70
    using
    0.68
    A
    0.65
    The
    0.63
    id
    0.63
    新的
    0.63
    است
    0.63
    <!
    0.62
     new
    0.61
    新しい
    0.61
    Act Density 0.590%

    No Known Activations